From owner-freebsd-stable Wed Mar 22 15:18:29 2000 Delivered-To: freebsd-stable@freebsd.org Received: from relay.ipf.net (relay.gigabell.net [195.211.211.23]) by hub.freebsd.org (Postfix) with SMTP id 64DBF37B586 for ; Wed, 22 Mar 2000 15:18:24 -0800 (PST) (envelope-from jachmann@gigabell.net) Received: (qmail 94383 invoked from network); 22 Mar 2000 23:18:22 -0000 Received: from habbib.bauernstand.org (HELO gigabell.net) (195.88.64.7) by relay.gigabell.net with SMTP; 22 Mar 2000 23:18:22 -0000 Message-ID: <38D954BC.4D49C9FA@gigabell.net> Date: Thu, 23 Mar 2000 00:18:20 +0100 From: Christian Jachmann Organization: Gigabell AG X-Mailer: Mozilla 4.07 [en] (X11; I; Linux 2.2.2 i586) MIME-Version: 1.0 To: Carroll Kong Cc: "stable@freebsd.org" Subject: Re: 3.4-Stable crashes..(heavy diskio+networking) References: <4.2.2.20000322175820.00e0e8c0@email.eden.rutgers.edu> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-stable@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG Carroll Kong wrote: > > At 11:27 PM 3/22/00 +0100, Christian Jachmann wrote: > >Hi, > > > >I'm running 3.4 Stable here, machine is a central newsfeed. > >Machine is running INN 2.2.2 (latest Snapshot) > >Network thrououtput is several hundred Gigabytes a Day. > >We do have about 150 streaming connections on this site. > > > >It reboots about every 24 hours. > > > >Is it bad ram ? > >or the temperature ? > >Maybe someone give me a hint. > > Ok. Your Maxusers seems a bit high. However, I have heard from several > people (it was not really too decisive even after looking at the archives) > that this is ok. How long has the system been working for? Is this It has been working in 2.2.* Release for about half a year, whithout any problem. After upgrading to 3.* (now 3.4) we got those crashes.... > problem just recent? Did it just occur after you inserted some new > ram. The error was a page fault. My guess is ram, or swap disk. (I could > not tell if you were swapping or not). What brand name RAM? How old is > the ram? My guess is heavily on the ram. I highly doubt if the program System is not swaping. I'll change the ram soon. But something interesting is in the gdb output : Fatal trap 12: page fault while in kernel mode fault virtual address = 0x12 fault code = supervisor read, page not present instruction pointer = 0x8:0xc0148ff0 stack pointer = 0x10:0xda886dac frame pointer = 0x10:0xda886dc8 code segment = base 0x0, limit 0xfffff, type 0x1b = DPL 0, pres 1, def32 1, gran 1 processor eflags = interrupt enabled, resume, IOPL = 0 current process = 2542 (innfeed) interrupt mask = trap number = 12 panic: page fault syncing disks... Fatal trap 12: page fault while in kernel mode fault virtual address = 0x12 fault code = supervisor read, page not present instruction pointer = 0x8:0xc0148ff0 stack pointer = 0x10:0xda8868dc frame pointer = 0x10:0xda8868f8 code segment = base 0x0, limit 0xfffff, type 0x1b = DPL 0, pres 1, def32 1, gran 1 processor eflags = interrupt enabled, resume, IOPL = 0 current process = 2542 (innfeed) interrupt mask = trap number = 12 panic: page fault --- looks like I got two page faults. same is also in the backtrace: #0 boot (howto=260) at ../../kern/kern_shutdown.c:285 #1 0xc0132b1d in panic (fmt=0xc01d78d2 "page fault") at ../../kern/kern_shutdown.c:446 #2 0xc01b66b2 in trap_fatal (frame=0xda8868a0, eva=18) at ../../i386/i386/trap.c:942 #3 0xc01b636b in trap_pfault (frame=0xda8868a0, usermode=0, eva=18) at ../../i386/i386/trap.c:835 #4 0xc01b5fca in trap (frame={tf_es = -1072234480, tf_ds = -981401584, tf_edi = 0, tf_esi = 0, tf_ebp = -628594440, tf_isp = -628594488, tf_ebx = 0, tf_edx = -2147483648, tf_ecx = 40, tf_eax = 40, tf_trapno = 12, tf_err = 0, tf_eip = -1072394256, tf_cs = 8, tf_eflags = 66118, tf_esp = 0, tf_ss = -1073610752}) at ../../i386/i386/trap.c:437 #5 0xc0148ff0 in m_copym (m=0x0, off0=0, len=46, wait=1) at ../../kern/uipc_mbuf.c:433 #6 0xc01747d2 in tcp_output (tp=0xd9e4ff00) at ../../netinet/tcp_output.c:505 and somewhat before: #17 0xc0159518 in sync (p=0xc0204318, uap=0x0) at ../../kern/vfs_syscalls.c:548 #18 0xc01326c5 in boot (howto=256) at ../../kern/kern_shutdown.c:203 #19 0xc0132b1d in panic (fmt=0xc01d78d2 "page fault") at ../../kern/kern_shutdown.c:446 #20 0xc01b66b2 in trap_fatal (frame=0xda886d70, eva=18) at ../../i386/i386/trap.c:942 #21 0xc01b636b in trap_pfault (frame=0xda886d70, usermode=0, eva=18) at ../../i386/i386/trap.c:835 #22 0xc01b5fca in trap (frame={tf_es = -980418544, tf_ds = -980418544, tf_edi = 0, tf_esi = 0, tf_ebp = -628593208, tf_isp = -628593256, tf_ebx = 0, tf_edx = 0, tf_ecx = 40, tf_eax = 40, tf_trapno = 12, tf_err = 0, tf_eip = -1072394256, tf_cs = 8, tf_eflags = 66118, tf_esp = 0, tf_ss = -1073610752}) at ../../i386/i386/trap.c:437 #23 0xc0148ff0 in m_copym (m=0x0, off0=0, len=46, wait=1) at ../../kern/uipc_mbuf.c:433 #24 0xc01747d2 in tcp_output (tp=0xd9e4ff00) at ../../netinet/tcp_output.c:505 #25 0xc017608a in tcp_usr_send (so=0xd97e7f00, flags=0, m=0xc188b800, nam=0x0, a second page fault.... Strage ? BYe -- Christian Jachmann Gigabell AG - Technik ++49 69 17084-0 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message